Theoretical models have shown that for populations in decline as a result of an introduced stressor, evolutionary rescue via natural selection can lead to renewed population growth, but if population size is reduced below a critical threshold, the population is at risk of extinction as a result of demographic stochasticity. While there is evidence for this theory to be applicable to a variety of stressors, we lack demonstration of its applicability in naturally occurring host-pathogen systems. We ask in our study what conditions are necessary for evolutionary rescue from extinction to be possible in host-pathogen systems. We begin by allowing evolution of resistance in host populations to occur in an ecologically-validated stochastic model of prairie dogs and the plague-causing bacterium Yersinia pestis, highlighting the rate of resistance development and variation of this rate (proxy for genetic variance) that would be necessary for prairie dogs to be evolutionarily rescued from plague-induced extinction. We then break the plague model down to compare different compartmental disease models, each with incorporated evolution of host resistance, to explore the ecological system components that allow evolutionary rescue from extinction to be a possibility.
Results/Conclusions
In simulated plague scenarios, there are only small ranges of population average initial resistance rate and rate variance that make evolutionary rescue from extinction possible. The population’s average resistance rate at the simulation’s start must be low enough that extinction is possible, but the population must not be so maladapted that extinction is inevitable. Similarly, the variance of this resistance rate must be high enough that an adaptive response is possible, but not so high as to result in a large well-adapted subpopulation that negates extinction. In exploring compartmental models for similar trends in potential for evolutionary rescue, we find that a SIR compartmental model and many variations thereof lack the potential for evolutionary rescue because the highly pathogenic disease runs its course too quickly. Introducing an infectious reservoir, as in the plague system, results in a constant selective pressure that prevents population regrowth, resulting in extinction. The conditions that allow evolutionary rescue in disease systems are a unique combination of sustained, highly pathogenic selective pressure balanced with nonlinear population growth that improves population survival, with moderate population adaptation and variance prior to infection. Thus, we believe evolutionary rescue via natural selection in host-pathogen systems to be a rare event.